CN106873142A - High-quality image acquisition device and method of tubercle bacillus detector - Google Patents
High-quality image acquisition device and method of tubercle bacillus detector Download PDFInfo
- Publication number
- CN106873142A CN106873142A CN201710153435.3A CN201710153435A CN106873142A CN 106873142 A CN106873142 A CN 106873142A CN 201710153435 A CN201710153435 A CN 201710153435A CN 106873142 A CN106873142 A CN 106873142A
- Authority
- CN
- China
- Prior art keywords
- focusing
- xyzz
- platforms
- image
- slide
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 40
- 241000193830 Bacillus <bacterium> Species 0.000 title claims abstract description 33
- 230000033001 locomotion Effects 0.000 claims abstract description 31
- 238000011156 evaluation Methods 0.000 claims description 45
- 230000000007 visual effect Effects 0.000 claims description 12
- 239000011159 matrix material Substances 0.000 claims description 8
- 230000001133 acceleration Effects 0.000 claims description 6
- 230000005540 biological transmission Effects 0.000 claims description 6
- 238000013519 translation Methods 0.000 claims description 5
- 230000000694 effects Effects 0.000 claims description 4
- 238000012935 Averaging Methods 0.000 claims description 2
- 238000004364 calculation method Methods 0.000 claims 1
- 206010036790 Productive cough Diseases 0.000 abstract description 7
- 210000003802 sputum Anatomy 0.000 abstract description 7
- 208000024794 sputum Diseases 0.000 abstract description 7
- 238000003384 imaging method Methods 0.000 abstract description 3
- 230000008859 change Effects 0.000 description 6
- 238000001514 detection method Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 6
- 239000000463 material Substances 0.000 description 4
- 230000001429 stepping effect Effects 0.000 description 4
- 201000008827 tuberculosis Diseases 0.000 description 4
- 238000011882 arthroplasty Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 3
- 208000035473 Communicable disease Diseases 0.000 description 2
- 230000009194 climbing Effects 0.000 description 2
- 230000005611 electricity Effects 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- NJPPVKZQTLUDBO-UHFFFAOYSA-N novaluron Chemical compound C1=C(Cl)C(OC(F)(F)C(OC(F)(F)F)F)=CC=C1NC(=O)NC(=O)C1=C(F)C=CC=C1F NJPPVKZQTLUDBO-UHFFFAOYSA-N 0.000 description 2
- 230000000737 periodic effect Effects 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 241000187479 Mycobacterium tuberculosis Species 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 230000001684 chronic effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 238000000205 computational method Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 238000013399 early diagnosis Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 208000015181 infectious disease Diseases 0.000 description 1
- 238000001000 micrograph Methods 0.000 description 1
- 238000000386 microscopy Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 244000052769 pathogen Species 0.000 description 1
- 230000001717 pathogenic effect Effects 0.000 description 1
- 230000001575 pathological effect Effects 0.000 description 1
- 230000002035 prolonged effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000004083 survival effect Effects 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G02—OPTICS
- G02B—OPTICAL ELEMENTS, SYSTEMS OR APPARATUS
- G02B21/00—Microscopes
- G02B21/24—Base structure
- G02B21/241—Devices for focusing
- G02B21/244—Devices for focusing using image analysis techniques
-
- G—PHYSICS
- G02—OPTICS
- G02B—OPTICAL ELEMENTS, SYSTEMS OR APPARATUS
- G02B21/00—Microscopes
- G02B21/24—Base structure
- G02B21/241—Devices for focusing
- G02B21/242—Devices for focusing with coarse and fine adjustment mechanism
Landscapes
- Physics & Mathematics (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- General Physics & Mathematics (AREA)
- Optics & Photonics (AREA)
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Microscoopes, Condenser (AREA)
Abstract
A high-quality image acquisition device and method for a tubercle bacillus detector are disclosed, wherein the device comprises an XYZz platform, a microscope, an industrial camera and a computer, and the method comprises initial setting, dynamic exposure strategy, XYZz platform motion resetting, primary focusing and scanning focusing. According to the method, the influence of uneven smearing of the tubercle bacillus sputum smear or other samples on image acquisition and imaging use is reduced through an image acquisition dynamic exposure strategy; the speed and robustness of the automatic focusing process are improved through dynamic focusing; by continuous focusing, the automatic focusing speed and the automation degree are greatly improved, and multi-view continuous automatic focusing can be realized; and the S-shaped curve and the PID algorithm motion model are utilized to reduce the motion error and improve the accuracy of focal plane positioning.
Description
Technical field
The invention belongs to technical field of machine vision, it is related to a kind of microscope Techniques of Automatic Focusing, more particularly to a kind of knot
The microscope Techniques of Automatic Focusing of core bacillus detector.
Background technology
Mycobacterium tuberculosis, is commonly called as tubercle bacillus, is to cause chronic infectious disease pathogen lungy.In the world
Interior, tuberculosis is still one of Infectious Diseases, has about 900 million news to send out patient and 2,000,000 Died Patients every year.Research table
Bright, early diagnosis, early treatment are to prevent and treat tuberculosis, improve the most effective means of tuberculosis patient survival rate.
From for pathologic angle, most reliable is also that to make a definite diagnosis the most important method of tuberculosis at present be tubercle bacillus microscopy
Method.Cervical arthroplasty method has that some can not ignore.First, cervical arthroplasty method needs pathologist or by professional training
Doctor operated.Secondly, cervical arthroplasty method workload is huge, one slide (sample) of complete interpretation need to be in microscope
At least 300 visuals field of lower observation.Again, manual detection cannot provide accurate quantitative information, such as tubercle bacillus concentration, form
Feature distribution etc., and cannot effectively record and check.Tubercle bacillus detector effectively overcomes artificial mirror by NI Vision Builder for Automated Inspection
The problem and shortage part that inspection method is present, realizes the automation and intellectuality of tubercle bacillus detection, will thoroughly overturn existing knot
Core disease diagnostic mode.
The acquisition of high-quality micro-image is closed the most in tubercle bacillus detector or even microscope automated detection system
One of key technology.Micro-image obtains system mainly includes microscope optical system, automatic focus movement platform and IMAQ
The part of system three constitutes.Wherein, microscope Techniques of Automatic Focusing is the basis and core technology that high quality graphic is obtained.It is existing
Microscope Techniques of Automatic Focusing is divided to and is broadly divided into two classes：Actively focus on and passive focusing.Actively focus on and pass through laser or infrared ray
Etc. the distance between mode measuring microscope object lens and target (sample focal plane) and it is adjusted, reaches focal position and complete
Focus on.Passive focusing is based on present image information, and state evaluation is focused to present image information using autofocus evaluation function,
Distance is accurate between Utilization assessment result and focal plane search strategy (focusing strategy) controlled motor adjustment micro objective and target
Focal plane position is navigated to, that is, completes to focus on.Actively focus on generally require addition laser or infrared distance measuring element, practicality compared with
It is low and relatively costly, in the automatic focus area of microscope and do not apply to, therefore be generally focused using the passive type of focusing.
Three aspects are included based on the passive microscope focusing technology key technology for focusing on：Autofocus evaluation function, focal plane
Search strategy and precision controlling.
In terms of autofocus evaluation function：Existing autofocus evaluation function studies comparative maturity, but for tubercle bacillus detection side
Face shows poor.Tubercle bacillus detection sample is Sputum smears, and smearing non-uniform areas, these areas are often contained in actual sample
Domain smearing thin and thick differs causes the image effective information fluctuation in microscope imaging very big.Sample (tubercle bacillus Sputum smears) is applied
The field color for smearing thickness is profound, exposure parameter is not changed and often leads to be imaged undesirable, the use to follow-up focusing completion image
Produce considerable influence.The thin field color of sample smearing is shallower, the autofocus evaluation function under the influence of picture noise and overexposure
Accuracy reduction, and then influence focusing strategy to cause focusing inaccurate or focus on failure.
In terms of the search strategy of focal plane：Existing focusing strategy technology includes blind person's climbing method and curve-fitting method two
Class.Blind person's climbing method arbitrarily chooses a focus direction first, and calculating this interval inner focusing according to the step distance of setting comments
The situation of change of valency function judges next step step direction.Because autofocus evaluation function is with unimodality, therefore in focal plane (ripple
Peak) place left side autofocus evaluation function monotonic increase, right side monotone decreasing.Conclude when crest is moved through and find focal plane, reversely
Crest is moved to complete to focus on.The method is more general, but because step distance is fixed, depth of field very little, required precision compared with
Tubercle bacillus high can cause small step-length, the situation of multiple stepping in focusing on automatically, less efficient.Curve-fitting method is burnt in distance
Plane farther out when due to evaluation function curve fluctuation very little its fitting accuracy it is very poor.
In focusing accuracy control aspect：Often there is kinematic error in actual focusing, such as motor movement is lost step, communication and prolonged
When, prior art does not relate to or provides associated solutions.
The content of the invention
The purpose of the present invention is achieved through the following technical solutions.
A kind of high quality graphic acquisition device of tubercle bacillus detector, including XYZz platforms, microscope, industrial camera,
Computer, it is characterised in that
The XYZz platforms include X layers of XYZz platforms, Y layers of XYZz platforms, XYZz platform bases, also including multiple steppings electricity
Machine, multiple grating scales and multigroup limit switch；
The microscope includes micro objective, slide, microscopic transmission light source, wherein, slide top is micro-
Endoscope objective lenses, lower section is microscopic transmission light source, and micro objective is used to obtain the micro-image of sample on slide；
The image that industrial camera is collected is the micro-image that micro objective obtains sample on slide, the micrograph
The computer is transferred to through capture card as after, the computer causes the stepping electricity on the XYZz platforms according to result
Machine, grating scale and limit switch collective effect, realize focusing on.
Preferably, stepper motor is four, and grating scale is three, and limit switch is three groups.
A kind of high quality graphic acquisition methods of tubercle bacillus detector, it uses above-mentioned device, it is characterised in that should
Method includes：
Part I, the slide containing sample to be tested is put into the XYZz platforms, the stepper motor of Z axis drives microscope
The absolute top of slide is moved to after thick quasi- Jiao's downward spiral rotation touching is spacing, IMAQ and dynamic exposure method, Z is opened
The stepper motor of axle drives the thin quasi- Jiao's spiral of microscope to scan for focusing on；
After the completion of Part II, the search are focused on, the stepper motor of the X-axis of the XYZz platforms, the stepper motor of Y-axis
Drive X layers of the XYZz platforms, Y layers of the XYZz platforms with S type curve movements, micro-image treats that focusing is regarded into next
Open country, is scanned focusing；
Part III, repetition Part II are until complete all automatic focusing tasks in the target visual field, focusing during this period is lost
Small range secondary focusing is carried out when losing or focusing on.
Preferably, fail or as needed, carry out small range secondary focusing if focused on.
Preferably, Part I includes：
Step one：Initial setting up；Specified location including the slide containing sample to be tested to be put into the XYZz platforms, beats
Industrial camera is opened, image collecting function is opened, camera parameter resets；
Step 2：Dynamic exposure strategy；The control industrial camera time for exposure obtains image averaging gray value and mean flow rate
Value, focuses on convolution matrix and intensity focus convolution matrix is multiplied with gray scale respectively, obtains exposing evaluation function after summation, is closed
The reason time for exposure；
Step 3：The motion of XYZz platforms resets；Realize the micro objective in Jiao including being moved by pid control algorithm
It is above plane is absolute and as small as possible with focal plane distance；
Step 4：It is first to focus on；Focal plane position p is obtained, the stepper motor of Z axis is moved to burnt putting down with S type acceleration curves
Face position p, completes search and focuses on, and preserves image.
Preferably, Part II includes：
Step 5：It is scanning focused；The stepper motor of X-axis, the stepper motor of Y-axis drive X layers of XYZz platforms, XYZz to put down respectively
Platform Y layers of translation, it is scanning focused to be recycled to the whole slide visual field to be measured；
Step 6：All automatic focusing is terminated in the whole slide visual field to be measured.
Preferably, Part III includes：Step 7：Secondary focusing, the secondary focusing is identical with scanning focused step, warp
Test 1/2nd of the empirical value that threshold value is scanning focused middle selection.
Present invention mainly solves the technical problem that high-quality micro-image is obtained, particularly tubercle bacillus detector high-quality
The technical problem that image is obtained, its major advantage is：
1. IMAQ dynamic exposure strategy of the invention, reduces tubercle bacillus Sputum smears or other samples smears uneven
It is even image to be obtained and is imaged the influence for using；
2. the dynamic focusing method that the present invention is used, especially focuses on the slide scanning focused stage using not on a large scale
The same type of focusing, improves the speed and robustness of automatic focusing；
3. the present invention uses sequential focusing, i.e., be made up of four stepper motors, three groups of limit switches, three grating scales
" XYZz " axle focuses on platform and automatic focusing speed and automaticity is greatly improved automatically, is capable of achieving many visuals field and continuously gathers automatically
It is burnt；
4. the high accuracy focus movement that the present invention is realized, by changing motor movement pattern, is calculated using S types curve and PID
Method motion model reduction kinematic error, improves the accuracy of locating focal plane.
Brief description of the drawings
By reading the detailed description of hereafter preferred embodiment, various other advantages and benefit is common for this area
Technical staff will be clear understanding.Accompanying drawing is only used for showing the purpose of preferred embodiment, and is not considered as to the present invention
Limitation.And in whole accompanying drawing, identical part is denoted by the same reference numerals.In the accompanying drawings：
Accompanying drawing 1 shows a kind of structure of the tubercle bacillus high quality graphic acquisition device according to embodiment of the present invention
Figure.
Accompanying drawing 2 to show and focusing on rank according to a kind of tubercle bacillus high quality graphic acquisition device of embodiment of the present invention
The focusing curve figure of section.
Accompanying drawing 3 shows poly- in scanning according to a kind of tubercle bacillus high quality graphic acquisition device of embodiment of the present invention
The autofocus logic figure in burnt stage.
Accompanying drawing 4 shows poly- in scanning according to a kind of tubercle bacillus high quality graphic acquisition device of embodiment of the present invention
The movement locus figure in burnt stage.
Specific embodiment
The illustrative embodiments of the disclosure are more fully described below with reference to accompanying drawings.Although showing this public affairs in accompanying drawing
The illustrative embodiments opened, it being understood, however, that may be realized in various forms the disclosure without the reality that should be illustrated here
The mode of applying is limited.Conversely, there is provided these implementation methods are able to be best understood from the disclosure, and can be by this public affairs
The scope opened it is complete convey to those skilled in the art.
According to the embodiment of the present invention, a kind of tubercle bacillus high quality graphic acquisition device structure is proposed, it is mainly wrapped
Include：XYZz platforms, microscope, industrial camera, computer and related electric equipment.As shown in figure 1,1 is micro objective, 2 are
Slide to be measured, 3 is X layers of XYZz platforms, and 4 is Y layers of XYZz platforms, 5 is XYZz platform bases.X layers and XYZz of XYZz platforms are put down
Platform Y layers is connected by two groups of straight-line guide rail slide blocks on Y layers, and Y layers of XYZz platforms and XYZz platform bases pass through
Two groups of straight-line guide rail slide blocks on pedestal are connected, and slide pushes X layers of center by the X layers of groove of one end,
Micro objective bottom and the initial spacing 3mm of slide surface.
In use, the distance between the adjustment bottom of micro objective 1 and the surface of slide 2 size, i.e., it is automatic poly-
Burnt performance directly perceived.X layers of XYZz platforms can be moved horizontally along " X-axis " direction of setting, and two ends are provided with limit switch.Y layers of motion
Form, stroke are identical with X layers, and direction is orthogonal.
Slide top is micro objective, and lower section is microscopic transmission light source, and the image that industrial camera is collected shows
Objective obtains the micro-image of sample on slide, and acquired card is transferred to computer.Computer is sent out according to result
Corresponding sports are sent to instruct, the motor, grating scale and limit switch collective effect on XYZz platforms realize that platform relative motion is instructed.
According to the embodiment of the present invention, realize finishing using above-mentioned tubercle bacillus high quality graphic acquisition device structure
Core bacillus high-quality micro-image acquisition methods, the method mainly includes following three part：
Part I, the slide containing sample to be tested is fixed and is put into " XYZz platforms ", and " Z axis " are that motor drives microscope
The absolute top of slide is moved to after thick quasi- Jiao's downward spiral rotation touching is spacing, IMAQ and dynamic exposure method is opened,
" z-axis " is that motor drives the thin quasi- Jiao's spiral of microscope to scan for focusing on.
Part II, " X-axis ", " Y " axle of " XYZz platforms " drives " X layers ", " Y of XYZz platforms after the completion of search focusing
It is mobile that layer " follows " curvilinear motion of S types ".Micro-image treats focus field into next, is scanned focusing.
Part III, repeats Part II until completing all automatic focusing tasks in the target visual field, focuses on lose during this period
Small range secondary focusing is carried out when losing or focusing on.
According to another implementation of the invention, comprise the following steps that：
Step one：Slide is put into, specified location is pushed into.
Industrial camera is opened, image collecting function is opened, camera parameter resets；
Step 2：Dynamic exposure strategy.
It is specifically included：Aperture size and exposure gain are not changed, the control industrial camera time for exposure realizes regulation collection
Mean picture brightness.This strategy starts in the starting of each focusing period or motor periodic motion starting.By obtaining 5 differences
The micro-image of time for exposure, using the functional relation between least square fitting time for exposure and mean picture brightness value,
Control image average brightness value is interval interior in [50,200], the as reasonable time for exposure；
Further, mean flow rate computational methods are as follows：
It is the congruent rectangle P of N number of pixel that testing image is divided into 9 width for M, highlyi, i ∈ [0,9].Rectangle
In (m, n) individual pixel gray value be g (m, n), brightness value be I (m, n).Homography PiAverage gray value beIt is flat
Equal brightness value isi∈[0,9]。
Image is transformed into the HSI color spaces for meeting eye-observation logic by RGB color, is made with luminance component I
It is the basic index of evaluation function.
G (m, n)=0.299Rm,n+0.587Gm,n+0.114Bm,n (1)
Wherein, Rm,n、Gm,n、Bm,nIt is the value of (m, n) pixel red, green, blue passage in RGB color in image.
By (1) formula and (2) Shi Ke get：
According to practical engineering experience, whenWhen very big or minimum, target focus features disappear, focusing error drastically increases
Greatly, should now be made by adjusting the time for exposureReturn normal value.
Finally, exposure convolution matrix is added, exposure evaluation function E is comprehensively drawn：
Wherein W, L are that gray scale focuses on convolution matrix and intensity focus convolution matrix, and the foundation microscope imaging depth of field is smaller, side
Edge gray value gradient is slightly less than middle section, sets convolution matrix to obtain more accurately exposing evaluation function.
Step 3：The motion of " XYZz " platform resets.
Coordinate grating scale to be rotated forward with pid control algorithm after X-axis and Y-axis reversion touching limit switch, make object lens (oil
Mirror) original position in detection.Z axis stepper motor drives the thick quasi- Jiao's downward spiral of microscope to rotate, by touching spacing opening
Grating scale is coordinated to be moved upwards with pid control algorithm after closing back to zero, to slide away from micro objective (oil mirror) top.This step
Rapid being moved by pid control algorithm realizes object lens in focal plane definitely top and with focal plane apart from as small as possible.
Wherein, include with pid control algorithm：Host computer sends moving target position p by CAN protocoldIt is to control
System, the motor current location p that control system periodically feeds back grating scalecWith target location pdDifference pdInput PID control is calculated
In method, calculating next cycle motor needs pulsed quantity and the controlled motor motion of motion, and multiple cycles are until motor movement repeatedly
To target location pd., with the system, kinematic accuracy can be significantly improved by by pid control algorithm.
Step 4：First focussing stage.
Z-axis stepper motor drives the thin quasi- Jiao's downward spiral uniform rotation of microscope, starts based on " blind person climbs the mountain " algorithm
Big step length searching is focused on.This search focus method is different from conventional blind person's hill-climbing algorithm, in each step (or each field of search
Between) starting enables dynamic exposure strategy and obtains optimum exposure image, and step-length (region of search) is more than conventional hill-climbing algorithm, focuses on
Accuracy is higher, speed faster.Autofocus logic is crossed and use after crest that " s types accelerate bent as shown in Fig. 2 crest is focal plane
Line " is accurately positioned to crest location and completes search focusing.
Further, according to another embodiment of the invention, first focus steps include：
1) z-axis motor is at the uniform velocity rotated forward (XYZz platforms are moved down), and object lens (oil mirror) reach original position.
2) industrial camera is focused evaluation per two field picture, until completing a region of search.The focusing evaluation in this stage
Function improves computational efficiency using the Std evaluation functions after improving.Such as formula 8.
Wherein,M, N are image pixel width and height.
3) autofocus evaluation function result is returned with grating scale current location, three values of average gray value are associated and are stored in two
In dimension group R [3, S], S rows, 3 row.
4) maximum autofocus evaluation function value V (i.e. interval interior all V in R [3, S] are found outstdIn maximum) and maximum it is flat
Grating scale position where equal gray value G, V is that this focuses on interval most picture rich in detail position, referred to as local optimum positions p.p
The average gray value of position, focusing evaluation of estimate are optimum gradation value g, optimum focusing value v.General, V=v, g ≠ G.
5) according to motor speed, grating scale resolution ratio, the thin quasi- Jiao's spiral resolution ratio of microscope etc. when judging focusing results
Factor adds boundary threshold, and this serial threshold value is empirical value.Moving boundaries threshold value T can be setm=0.02, exposure boundaries threshold value Te=
0.1 and evaluation boundary threshold Tf=0.03；
If p >=STm&p≤S·(1-Tm)；
I.e. local optimum positions fall in ideal interval, judge that p is focal plane position；
If p ＜ STm＆ | g-G | ＜ GTe；
I.e. local optimum positions judge that this interval focuses on failure in interval left end and picture material without significant change, return
Initial position is focused on again；
If p ＞ S (1-Tm) ＆ | g-G | ＜ GTe；
I.e. local optimum positions judge that this interval, without focal plane, is continued in interval right-hand member and picture material without significant change
Search is focused on；
If p ＞ S (1-Tm) ＆ | g-G | ＞ GTe；
I.e. local optimum positions are in interval right-hand member and picture material has significant change, judge that this interval, without focal plane, is continued
Search is focused on；
If p ＜ S (1-Tm) ＆ | g-G | ＞ GTe；
I.e. local optimum positions are in interval left end and picture material has significant change, judge that this interval focuses on failure, return
Initial position is reversely focused on；
6) motor " s types acceleration curve " moves to focal plane position p, completes search and focuses on, and preserves image；
7) focal plane position image is focused evaluation, autofocus evaluation function value v ' again.IfJudge
It is accurate to focus on；Failure is otherwise focused on, secondary focusing is carried out.
Step 5：It is scanning focused.
X-axis, y-axis motor drive X layers, Y layers of platform translation, it then follows predetermined movement track is waited to focus on micro- regarding into next
Open country, movement locus is as shown in Figure 3.Z-axis rotates backward fixed range and reaches scanning focused original position.Aggregation is opened to evaluate and oneself
Dynamic exposure strategies, z-axis rotates forward 2 times of fixed ranges, and this interval optimum focusing evaluation of estimate position is focal plane position.It is poly-
Burnt logic is as shown in figure 4, scanning focused be recycled to the whole slide visual field to be measured all automatic focusing is terminated.
Further, according to another embodiment of the invention, scanning focused step is as follows：
1) it is scanning focused sweep radius to take empirical value r, typically according to sample to be tested focal plane depth of field size, stepping
Motor step-length, grating scale subdivision are chosen；
2) z-axis stepper motor drives thin quasi- Jiao's spiral " S types curve movement " of microscope to rotate backward r distances；
3) z-axis stepper motor drives the thin quasi- Jiao's spiral of microscope at the uniform velocity to rotate forward 2r distances, is evaluated using Tenegrad
Function pair is focused evaluation per two field picture, until completing a sweep interval；
4) autofocus evaluation function result is returned with grating scale current location, three values of average gray value are associated and are stored in two
In dimension group R [3, S], S rows, 3 row；
5) maximum autofocus evaluation function value V is with grating scale position where maximum average gray value G, V in finding out R [3, S]
This focuses on interval most picture rich in detail position, referred to as local optimum positions p.The average gray value of p positions, focusing evaluation of estimate
It is optimum gradation value g, optimum focusing value v.General, V=v, g ≠ G.
6) motor " s types acceleration curve " moves to focal plane position p, completes scanning focused, preserves focal plane position
Image；
7) focal plane position image is focused evaluation, autofocus evaluation function value v ' again.IfJudge
It is accurate to focus on；Failure is otherwise focused on, secondary focusing is carried out.
Step 6：All automatic focusing is terminated in the whole slide visual field to be measured；
Step 7：Secondary focusing.Secondary focusing is identical with scanning focused step, and empirical value is scanning focused two/
One.
Specific embodiment：
According to another example of the invention, a kind of tubercle bacillus high quality graphic is obtained and device technique scheme is as follows：
The present embodiment includes Olympus BX43 microscopes, XYZz motion platforms, JAIGO-5000M-USB high-speed industrials
Camera, it is acid-fast stain tubercle bacillus Sputum smears that sample (slide) is focused on automatically.
XYZz platforms as shown in figure 1,1 be micro objective, 2 be slide to be measured, 3 be X layers of XYZz platforms, 4 is XYZz
Platform Y layers, 5 be XYZz platform bases.Platform X layers is connected solid with Y layers by two groups of straight-line guide rail slide blocks on Y layers
Fixed, Y layers is connected with platform base by two groups of straight-line guide rail slide blocks on pedestal, and slide is by X layers of one end
Groove pushes X layers of center, micro objective bottom and the initial spacing 3mm of slide surface.
X layers of XYZz platforms can be moved horizontally along " X-axis " direction of setting, and two ends are provided with limit switch.Y layers of forms of motion,
Stroke is identical with X layers, and direction is orthogonal.
Slide top is micro objective, and lower section is microscopic transmission light source, and the image that industrial camera is collected shows
Objective obtains the micro-image of sample on slide, and acquired card is transferred to computer.Computer is sent out according to result
Corresponding sports are sent to instruct, the motor, grating scale and limit switch collective effect on XYZz platforms realize that platform relative motion is instructed.
Comprise the following steps that：
Step one：Tubercle bacillus Sputum smears are put into XYZz platforms, center is pushed into micro objective position directly below.
Industrial camera is opened, image collecting function is opened, camera parameter is reset；
Step 2：Start dynamic exposure strategy.This strategy is in the starting of each focusing period or motor periodic motion starting
Start, it is 3000 μ s, 5000 μ s, 8000 μ s, 12000 μ s, the micro-image of 16000 μ s to obtain 5 time for exposure first, secondly
Using the functional relation between least square fitting time for exposure and mean picture brightness value val, mean picture brightness is controlled
Value is interval interior in [50,200], as reasonable time for exposure t；
Step 3：X-axis and Y-axis drive X layers, Y layers of platform reversion, coordinate grating scale to be controlled with PID after touching limit switch
Algorithm positive movement processed is located at the upper left corner of slide specimen to object lens, and this is the automatic starting point for focusing on.Z axis stepper motor band
Dynamic thick quasi- Jiao's spiral reversion (downward) of microscope, rotates forward (upward) to slide away from 100 times of things of microscope after touching limit switch
At mirror (oil mirror) 3mm；
Step 4：Search is focused on.Comprise the following steps that：
1) z-axis stepper motor drives thin quasi- Jiao's spiral forward direction uniform rotation (downward) of microscope, and move distance is 300.
2) industrial camera is focused evaluation per two field picture, until complete motor stopping.The autofocus evaluation function in this stage
Using the std evaluation functions after improvement, such as following formula.
Wherein,M, N are image pixel width and height.
3) autofocus evaluation function result is returned with grating scale current location, three values of average gray value are associated and are stored in two
In dimension group R [3, S], S rows, 3 row.
4) maximum autofocus evaluation function value V (i.e. interval interior all V in R [3, S] are found outstdIn maximum) and maximum it is flat
Grating scale position where equal gray value G, V is that this focuses on interval most picture rich in detail position, referred to as local optimum positions p.p
The average gray value of position, focusing evaluation of estimate are optimum gradation value g, optimum focusing value v.General, V=v, g ≠ G.
5) according to motor speed, grating scale resolution ratio, the thin quasi- Jiao's spiral resolution ratio of microscope etc. when judging focusing results
Factor adds boundary threshold, and this serial threshold value is empirical value.For tubercle bacillus micro-image, moving boundaries threshold value T can be setm=
0.02nd, exposure boundaries threshold value Te=0.1 and evaluation boundary threshold Tf=0.03；
If p >=STm&p≤S·(1-Tm)；
Judge that p is focal plane position；
If p ＜ STm＆ | g-G | ＜ GTe；
R [3, S] is emptied, and is returned to initial position and is focused on again；
If p ＞ S (1-Tm) ＆ | g-G | ＜ GTe；
R [3, S] is emptied, and continues search for focusing on；
If p ＞ S (1-Tm) ＆ | g-G | ＞ GTe；
R [3, S] is emptied, and continues search for focusing on；
If p ＜ S (1-Tm) ＆ | g-G | ＞ GTe；
Failure is focused on, R [3, S] is emptied, return to initial position and reversely focus on；
6) motor " s types acceleration curve " moves to focal plane position p, completes search and focuses on, and preserves image；
7) focal plane position image is focused evaluation, autofocus evaluation function value v ' again.IfJudge
It is accurate to focus on；Failure is otherwise focused on, secondary focusing is carried out.
Step 5：X-axis motor drives X layers of positive translation 200 of XYZz platforms, starts into next microscopic field to be measured
It is scanning focused.When object lens are located at sample to be tested X axis edge, X layers of stop motion.X layers is returned starting point, and y-axis motor drives
The positive translation 200 of Y layers of XYZz platforms, continues to scan on focusing.It is scanning focused to comprise the following steps that：
1) the tubercle bacillus Sputum smears depth of field is smaller, takes scanning focused sweep radius r=30；
2) z-axis stepper motor drives thin quasi- Jiao's spiral " S types curve movement " of microscope to rotate backward r distances.
3) z-axis stepper motor drives the thin quasi- Jiao's spiral of microscope at the uniform velocity to rotate forward 2r distances, is evaluated using Tenegrad
Function pair is focused evaluation per two field picture, until completing a sweep interval；
4) autofocus evaluation function result is returned with grating scale current location, three values of average gray value are associated and are stored in two
In dimension group R [3, S], S rows, 3 row.
5) maximum autofocus evaluation function value V is with grating scale position where maximum average gray value G, V in finding out R [3, S]
This focuses on interval most picture rich in detail position, referred to as local optimum positions p.The average gray value of p positions, focusing evaluation of estimate
It is optimum gradation value g, optimum focusing value v.General, V=v, g ≠ G.
6) motor " s types acceleration curve " moves to focal plane position p, completes scanning focused, preserves focal plane position
Image；
7) focal plane position image is focused evaluation, autofocus evaluation function value v ' again.IfJudge
It is accurate to focus on；Failure is otherwise focused on, secondary focusing is carried out.
Step 6：All automatic focusing is terminated in the whole slide visual field to be measured.
Step 7：Secondary focusing.Secondary focusing is identical with scanning focused step, empirical value r=15.
The above, the only present invention preferably specific embodiment, but protection scope of the present invention is not limited thereto,
Any one skilled in the art the invention discloses technical scope in, the change or replacement that can be readily occurred in,
Should all be included within the scope of the present invention.Therefore, protection scope of the present invention should be with the protection model of the claim
Enclose and be defined.
Claims (7)
1. a kind of high quality graphic acquisition device of tubercle bacillus detector, including XYZz platforms, microscope, industrial camera, meter
Calculation machine, it is characterised in that
The XYZz platforms include X layers of XYZz platforms, Y layers of XYZz platforms, XYZz platform bases, also including multiple stepper motors,
Multiple grating scales and multigroup limit switch；
The microscope includes micro objective, slide, microscopic transmission light source, wherein, slide top is microscope thing
Mirror, lower section is microscopic transmission light source, and micro objective is used to obtain the micro-image of sample on slide；
The image that industrial camera is collected is the micro-image that micro objective obtains sample on slide, after the micro-image
Acquired card is transferred to the computer, the computer caused according to result stepper motor on the XYZz platforms,
Grating scale and limit switch collective effect, realize focusing on.
2. a kind of device as claimed in claim 1, it is characterised in that stepper motor is four, grating scale is three, spacing to open
Guan Weisan groups.
3. a kind of high quality graphic acquisition methods of tubercle bacillus detector, it uses the device described in claim 1 or 2, its
It is characterised by, the method includes：
Part I, the slide containing sample to be tested is put into the XYZz platforms, the stepper motor of Z axis drives microscope slightly accurate
The absolute top of slide is moved to after burnt downward spiral rotation touching is spacing, IMAQ and dynamic exposure method is opened, Z axis
Stepper motor drives the thin quasi- Jiao's spiral of microscope to scan for focusing on；
After the completion of Part II, the search are focused on, the stepper motor of the X-axis of the XYZz platforms, the stepper motor of Y-axis drive
, with S type curve movements, micro-image is treated focus field, is entered into next for X layers of the XYZz platforms, Y layers of the XYZz platforms
Row is scanning focused；
Part III, repeat Part II until completing all automatic focusing tasks in the target visual field, focus on during this period failure or
Small range secondary focusing is carried out during focusing.
4. a kind of method as claimed in claim 3, it is characterised in that if focusing on failure or as needed, carry out small model
Enclose secondary focusing.
5. a kind of method as described in claim 3 or 4, it is characterised in that Part I includes：
Step one：Initial setting up；Specified location including the slide containing sample to be tested to be put into the XYZz platforms, opens work
Industry camera, opens image collecting function, and camera parameter resets；
Step 2：Dynamic exposure strategy；The control industrial camera time for exposure obtains image averaging gray value and average brightness value, point
Convolution matrix is not focused on gray scale and intensity focus convolution matrix is multiplied, obtain exposing evaluation function after summation, rationally exposed
The light time；
Step 3：The motion of XYZz platforms resets；Realize the micro objective in focal plane including being moved by pid control algorithm
It is above definitely and as small as possible with focal plane distance；
Step 4：It is first to focus on；Focal plane position p is obtained, the stepper motor of Z axis is moved to focal plane position with S type acceleration curves
P is put, search is completed and is focused on, preserve image.
6. a kind of method as described in claim 3 or 4 or 5, it is characterised in that Part II includes：
Step 5：It is scanning focused；The stepper motor of X-axis, the stepper motor of Y-axis drive X layers of XYZz platforms, XYZz platforms Y respectively
Layer translation, it is scanning focused to be recycled to the whole slide visual field to be measured；
Step 6：All automatic focusing is terminated in the whole slide visual field to be measured.
7. a kind of method as described in claim 3 or 4 or 5 or 6, it is characterised in that Part III includes：Step 7：It is secondary
Focus on, the secondary focusing is identical with scanning focused step, and empirical value is two points of the empirical value of scanning focused middle selection
One of.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710153435.3A CN106873142B (en) | 2017-03-15 | 2017-03-15 | High-quality image acquisition device and method of tubercle bacillus detector |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710153435.3A CN106873142B (en) | 2017-03-15 | 2017-03-15 | High-quality image acquisition device and method of tubercle bacillus detector |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106873142A true CN106873142A (en) | 2017-06-20 |
| CN106873142B CN106873142B (en) | 2019-04-12 |
Family
ID=59171779
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710153435.3A Expired - Fee Related CN106873142B (en) | 2017-03-15 | 2017-03-15 | High-quality image acquisition device and method of tubercle bacillus detector |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106873142B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107505698A (en) * | 2017-07-17 | 2017-12-22 | 中国家用电器研究院 | A kind of biological identification counting device and method |
| CN109491297A (en) * | 2018-12-12 | 2019-03-19 | 湖南爱威医疗科技有限公司 | Microscope photograph acquisition method and system |
| CN110992408A (en) * | 2019-11-18 | 2020-04-10 | 迪普派斯医疗科技(山东)有限公司 | Digital section processing method and system based on pathological microscope |
| CN111105346A (en) * | 2019-11-08 | 2020-05-05 | 同济大学 | Full-scanning microscopic image splicing method based on peak value search and gray template registration |
| CN111277744A (en) * | 2018-12-04 | 2020-06-12 | 武汉华大智造科技有限公司 | Focusing method and system, gene sequencer and computer readable storage medium |
| CN111474103A (en) * | 2019-01-24 | 2020-07-31 | 武汉兰丁医学高科技有限公司 | Automatic focusing scanning method and system for bone marrow cell glass slide |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2003063078A1 (en) * | 2002-01-21 | 2003-07-31 | Matsushita Ecology Systems Co., Ltd. | Imaging device |
| CN101629911A (en) * | 2009-08-21 | 2010-01-20 | 常州达奇信息科技有限公司 | Automatic tubercle bacillus detector |
| CN102304472A (en) * | 2011-05-18 | 2012-01-04 | 东华大学 | Device and method for detecting sputum smear tubercle bacillus quickly |
| CN202453578U (en) * | 2012-02-21 | 2012-09-26 | 上海北昂科学仪器有限公司 | Oil immersion lens-based automatic microscopic image scanning acquisition device |
| JP2014085599A (en) * | 2012-10-25 | 2014-05-12 | Olympus Corp | Microscope |
| WO2016101092A1 (en) * | 2014-12-22 | 2016-06-30 | 杭州唐光科技有限公司 | Digital microscope and focusing method thereof |
-
2017
- 2017-03-15 CN CN201710153435.3A patent/CN106873142B/en not_active Expired - Fee Related
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2003063078A1 (en) * | 2002-01-21 | 2003-07-31 | Matsushita Ecology Systems Co., Ltd. | Imaging device |
| CN101629911A (en) * | 2009-08-21 | 2010-01-20 | 常州达奇信息科技有限公司 | Automatic tubercle bacillus detector |
| CN102304472A (en) * | 2011-05-18 | 2012-01-04 | 东华大学 | Device and method for detecting sputum smear tubercle bacillus quickly |
| CN202453578U (en) * | 2012-02-21 | 2012-09-26 | 上海北昂科学仪器有限公司 | Oil immersion lens-based automatic microscopic image scanning acquisition device |
| JP2014085599A (en) * | 2012-10-25 | 2014-05-12 | Olympus Corp | Microscope |
| WO2016101092A1 (en) * | 2014-12-22 | 2016-06-30 | 杭州唐光科技有限公司 | Digital microscope and focusing method thereof |
Non-Patent Citations (2)
| Title |
|---|
| 张燕: "《医用形态实验学》", 31 December 2015 * |
| 黄金宝: "显微镜自动调光及聚焦算法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107505698A (en) * | 2017-07-17 | 2017-12-22 | 中国家用电器研究院 | A kind of biological identification counting device and method |
| CN107505698B (en) * | 2017-07-17 | 2024-01-12 | 中国家用电器研究院 | Biological identification counting device and method |
| CN111277744A (en) * | 2018-12-04 | 2020-06-12 | 武汉华大智造科技有限公司 | Focusing method and system, gene sequencer and computer readable storage medium |
| CN111277744B (en) * | 2018-12-04 | 2022-04-15 | 武汉华大智造科技有限公司 | Focusing method and system, gene sequencer and computer readable storage medium |
| CN109491297A (en) * | 2018-12-12 | 2019-03-19 | 湖南爱威医疗科技有限公司 | Microscope photograph acquisition method and system |
| CN111474103A (en) * | 2019-01-24 | 2020-07-31 | 武汉兰丁医学高科技有限公司 | Automatic focusing scanning method and system for bone marrow cell glass slide |
| CN111105346A (en) * | 2019-11-08 | 2020-05-05 | 同济大学 | Full-scanning microscopic image splicing method based on peak value search and gray template registration |
| CN110992408A (en) * | 2019-11-18 | 2020-04-10 | 迪普派斯医疗科技(山东)有限公司 | Digital section processing method and system based on pathological microscope |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106873142B (en) | 2019-04-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106873142B (en) | High-quality image acquisition device and method of tubercle bacillus detector | |
| CN105652429B (en) | A kind of microscopic cell glass slide scanning auto focusing method based on machine learning | |
| US7876948B2 (en) | System for creating microscopic digital montage images | |
| JP4917331B2 (en) | Image acquisition apparatus, image acquisition method, and image acquisition program | |
| US20120236120A1 (en) | Automatic stereological analysis of biological tissue including section thickness determination | |
| CN101441320B (en) | High dynamic image acquisition device based on microscopic imaging detection and method thereof | |
| CN100392403C (en) | Automatic counting method of leucocyte number in blood microimage | |
| KR101891364B1 (en) | Fast auto-focus in microscopic imaging | |
| CN109085695B (en) | Method for quickly focusing and photographing plane sample | |
| CN111007661A (en) | Microscopic image automatic focusing method and device based on deep learning | |
| CN106461928B (en) | Image processing apparatus, photographic device, microscopic system and image processing method | |
| CN109239900A (en) | A kind of full-automatic quick focusing method for the big visual field acquisition of microscopic digital image | |
| CN110146974A (en) | A kind of intelligent biological microscope | |
| EP3698193A1 (en) | Adaptive sensing based on depth | |
| JP7319903B2 (en) | High-speed TAG lens-assisted 3D measurement and extended depth-of-field imaging | |
| CN111220615A (en) | Inclined three-dimensional scanning microscopic imaging system and method | |
| CN109272575B (en) | Method for improving modeling speed of digital slice scanner | |
| CN109001902A (en) | Microscope focus method based on image co-registration | |
| CN106878617B (en) | A kind of focusing method and system | |
| CN109752831A (en) | A kind of microscope auto focusing method, system and device that focal time is controllable | |
| CN105651699B (en) | It is a kind of based on the dynamic of area array cameras with burnt method | |
| CN106556350B (en) | The measuring method and a kind of microscope of microscopic slide curved surface height value | |
| CN112069735B (en) | Full-slice digital imaging high-precision automatic focusing method based on asymmetric aberration | |
| US11256078B2 (en) | Continuous scanning for localization microscopy | |
| CN110490801A (en) | Metallographic microscope matrix super large picture high speed joining method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20190412 Termination date: 20210315 |